intersection point

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • intersection point

    Does anyone know a way to construct an intersection point between two parallel cylinders? Since pc-dmis wants to use the center line of both cylinders and the centerlines do not intersect, the math fails. Is there a way to get the point where the two outside surfaces of the cylinders intersect?

  • #2
    wouldnt that create a line and not a point if the cylinders are tangent to each other or 2 lines if they overlap.

    it is possible to do a workaournd using constructed features but need a bit more info
    are you looking for the intersection point at a specific level of the cylinder? are you looking at the cylinders along the cylinder axis, so basically seeing 2 circles? or are you looking for a 3d point in space?
    sigpic

    Comment


    • #3
      Post a print or this post isn't going to go anywhere

      Comment


      • #4
        Create a plane perpendicular to the cylinders at one end, construct circles by intersection cylinder-plane, create points by intersection circle1-circle2 (and the other way around to get the other intersection point). Do all of that at the other end of the cylinders too, and you have what's needed to create the intersection lines. Do note that this will not reflect form errors in the cylinders - no real hits are involved.
        Last edited by AndersI; 09-12-2019, 09:21 AM.
        AndersI
        SW support - Hexagon Metrology Nordic AB

        Comment


        • #5
          Yes, a line would be great too. But I’ve tried that, and the math fails on that as well . I’m looking along the axis of the cylinders . So yes, basically 2 circles . So what I’m trying to do , is get the upper intersection point where the cylinders overlap. So say that is in my +X axis. In my -X axis I have to get another symmetrical intersection point. Then I need to dimension the length between those two intersection points.

          Thanks for the Reply.
          Attached Files

          Comment


          • AndersI
            AndersI commented
            Editing a comment
            Sorry, I don't see two cylinders in your picture - you have to draw something a bit more obvious to those that don't have the part...

        • #6
          Well I’m talking about those two partial radiuses you see there in the picture. I was just measuring them as cylinders . If it would be easier to measure them both as circles I could do that. Try to imagine you’re looking right through that view and I’m measuring the circles or cylinders in the Z/X axis.

          Comment


          • AndersI
            AndersI commented
            Editing a comment
            The axes are not shown...

        • #7
          If you where the letter B is on that picture the line that comes right past it down to the part. That is the intersection point I’m looking to construct .

          Comment


          • #8
            All I see are linears... not even a radius call out.
            PcDmis 2015.1 SP10 CAD++
            Global 7-10-7 DC800S

            Comment


            • #9
              the only radius i see is the one marked in red. the only intersections i see are the 2 lines marked in yellow
              either way i'd intersect the line with the radius to create the intersection point to dimension from

              the bottom intersection appears to be tangent which can cause the math fail issue if the part radius as machined isnt actually tangent to the line. then things start getting creative

              realistically the radius location should be dimensioned to the radius center point not the intersetion/tangent point to other features
              image_17503.jpg
              Last edited by mckenzie; 09-12-2019, 10:09 AM.
              sigpic

              Comment


              • #10
                Originally posted by mckenzie View Post
                the only radius i see is the one marked in red. the only intersections i see are the 2 lines marked in yellow
                I guess it's the ellips that is his other 'cylinder', but it definitely isn't circular (in this view)…

                We'll have to wait for jefman's magic to solve this one…

                Edit: Some links - https://www.analyzemath.com/CircleEq...pse_inter.html, http://yehar.com/blog/?p=2926 (this one is cool...)
                Last edited by AndersI; 09-12-2019, 10:28 AM.
                AndersI
                SW support - Hexagon Metrology Nordic AB

                Comment


                • mckenzie
                  mckenzie commented
                  Editing a comment
                  im having a hard time wrapping my brain around the ellipse as a feature. if it was below the other geometry it would have hidden lines where it intersects thru the geometry and likewise if it was above the other geometry the ellipse would have some hidden lines. it looks like its just an ellipse drawn on the drawing. it's too early in the morning here, im confused

                • AndersI
                  AndersI commented
                  Editing a comment
                  Aren't we all?

                • mckenzie
                  mckenzie commented
                  Editing a comment
                  i love how we (qc folks) as a group tend to be the most confused group in the shop. the shop floor folks are never confused, ive watched them rapid the spindle -z through the part, workholding and into the table with such confidence.

                  just for the record i was a machinist before i got smart and moved into the air conditioning

              • #11
                This is the intersection point I’m looking for . And that bottom arc is a circular segment.
                Attached Files

                Comment


                • #12
                  so just to make sure i understand the blue line is a shallow arc segment and your looking for the intersection of the blue arc and the red arc? if so thats going to be tricky pcdmis doesnt handle short arc segments very well. trying to extrapolate a circle from such a small section of the circle, any slight deviation in the arc will have a huge affect on the extrapolated circle diameter and location. usually when the engineers create stuff like this the 2 arcs are tangent so any little deviation in the features as machined and the arcs wont read as intersecting. im hesitant to suggest a work around because it will just be an approximation. what is the tolerance for the liner dimension youre trying to measure?
                  at a minimum i would treat them as 2d features and create 2 circles and start there.

                  image_17503.jpg
                  sigpic

                  Comment


                  • #13
                    The tolerance for the linear dimension is +/- .010. I know small circular segments never tend to read right . ESP when you have a tiny piece of a 29.667 diameter . I guess I’m not really concerned about the dimension coming in because the chances are slim, but more concerned about if there IS a way to theoretically do this in pcdmis. I’ve ran Into this multiple times and it’s driving me nuts . Haha

                    Comment


                    • #14
                      Originally posted by Gagedude127 View Post
                      The tolerance for the linear dimension is +/- .010. I know small circular segments never tend to read right . ESP when you have a tiny piece of a 29.667 diameter . I guess I’m not really concerned about the dimension coming in because the chances are slim, but more concerned about if there IS a way to theoretically do this in pcdmis. I’ve ran Into this multiple times and it’s driving me nuts . Haha
                      Ok I get it now. It's possible though may not yield the best/actual results. Think 2D when working on this one. Create a plane normal to the drawing view and then construct 2D circles from your cylinders (assuming you have measured them based on your previous posts) and intersect a point that way. You might have to create 2D circles of your 2D circles to get them to "rotate" the correct way and get the point on the correct intersection. The last part might be confusing sorry I'm not at the CMM at the moment... but it's in the circle construction window somewhere, someone please correct me on this one.
                      PcDmis 2015.1 SP10 CAD++
                      Global 7-10-7 DC800S

                      Comment


                      • AndersI
                        AndersI commented
                        Editing a comment
                        Intersection circle/circle can produce both points without any extra circles. Just flip the order of the two circles in the construction command.

                      • Kp61dude!
                        Kp61dude! commented
                        Editing a comment
                        If you can achieve that by explained methods...awesome! I tried and don't get them. Running 2015.1

                    • #15
                      But sorry, yes, that is the point I’m looking for.

                      Comment

                      Related Topics

                      Collapse

                      Working...
                      X